A System for today. I wanted to walk you guys through what I was thinking and I want your feedback on whether or not this is viable or if maybe there is another way to filter it. Thanks in advance..
HYPOTHESIS: When do records begin to mean something? Let's say a team is 75% or 100%...well if it is the beginning of the season I don't think it is fair to say anyone is really bad or good; however, my thought today was this: during a point of the season where stats mean things due to sufficient sample sizes, does the public overrate anything? Of course they do; I picked the most square metric gamblers can look at: average points per game. Here's what I found:
In the middle of the season (Jan and Feb) where stats have good sample and become meaningful, since 2006, road teams averaging very few points per game (less than one standard deviation from the median) are 1218-1080-44 ATS (53%)!
*3 parameters (simple)
*2342 samples (massive)
*Logical? (I think so)
It gets better. If you think about what is going on here: the team is getting undervalued because they aren't putting up points...well that doesn't tell a full story like, were they injured in those games or were the up against very hard teams or something. It is just a square stat...teams between +10 and +23 points (undervalued) are 519-376-17 (58%) ATS.
Recap: big road underdogs averaging a "bad" number of points per game in the middle of the season are a great play on.
Active on: Fordham +16.5 and Marist +12.5
I am personally taking Fordham +16.5